HXC50136
Product Specifications
HXC50136
Hisiwell — Discrete & Power Semiconductors
Product Overview
The HXC50136 from Hisiwell is a discrete power semiconductor delivering efficient switching and power conversion. Hisiwell is a recognized semiconductor supplier providing reliable components for global electronics manufacturing.
Superb Automation partners with leading brands including Qorvo, TI, ADI, XILINX, NVIDIA, Intel, ST, MACOM, Renesas, and Coilcraft to provide complete BOM and supply-chain solutions.
Key Features
- Low Rds(on) for minimal conduction losses in power switching
- Fast switching speed with low gate charge for high-frequency operation
- Avalanche-rated for robust operation under inductive load conditions
- High junction temperature rating for demanding environments
- Logic-level gate drive compatible for simplified design
- RoHS and AEC-Q101 qualified options available
Key Specifications
| Vds / Vce | V |
| Id / Ic | A |
| Rds(on) / Vce(sat) | mΩ |
| Qg | nC |
| Technology | — |
| Package | — |
| Configuration | — |
| Mounting | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
HXC50136: 1000 W with 10 dB of gain, and the driver is the hard part
The HXC50136 is the largest row in the 960 to 1215 MHz group: 1000 W, internally matched, 50 V, 10 dB of gain, 50 per cent drain efficiency, marked Pulse on the index. The vendor page describes a 1000 W high performance internally matched LDMOS transistor for avionics at 960 to 1215 MHz, notes high power and high ruggedness, and lists one application, ISM RF energy. The data sheet link on the page points at the site root and serves nothing, and the parameters block is a picture, so no pulse width, duty cycle, thermal resistance or breakdown figure can be read in text.
At this power level the number that decides whether the part can be used is not the output. It is the gain. Ten decibels means the drive for 1000 W is 100 W, so the driver is a 100 W amplifier in its own right, with its own heat sink, its own bias and its own matching. A two stage chain on this row is a 50 dB chain built from two large parts, and the board area, the supply current and the cost of the driver are all set by that one figure on the index line.
The second number is the heat. At 50 per cent efficiency 1000 W of output means 1000 W of heat, and the only thing that makes that manageable is duty cycle. Since the vendor publishes neither pulse width nor duty, the heat sink cannot be sized from the page. That is a gap a purchase order should not cross.
The HXC50136 rows as the vendor publishes them
| Part class | LDMOS RF power transistor, internally matched |
| Output, index row | 1000 W |
| Output in dBm, calculated | 60.00 dBm |
| Frequency range, index row | 0.96 to 1.215 GHz |
| Frequency range, vendor page | 960 to 1215 MHz |
| Drain rail, index row | 50 V |
| Gain, index row | 10 dB |
| Drive required, calculated | 100 W, or 50.00 dBm |
| Drain efficiency, index row | 50 per cent |
| DC input at that efficiency, calculated | 2000 W, 40 A at 50 V |
| Heat to remove, calculated | 1000 W |
| Mode, index row | Pulse |
| Modes on the vendor page | Not stated |
| Application list, vendor page | ISM RF energy |
| Ruggedness statement | High ruggedness, no figure attached |
| Pulse width and duty limit | Not published |
| Peak output against average output | Not separated on either page |
| Thermal resistance, junction to case | Not published in text |
| Drain to source breakdown | Not published in text |
| Input, output and feedback capacitance | Not published in text |
| Load mismatch tolerance | Not published |
| Junction temperature maximum | Not published in text |
| Package outline | Not published in text |
| Data sheet link on the vendor page | Points at the site root, serves nothing |
| Parameters block on the vendor page | Published as a picture, not as text |
| Orderable addendum | None published |
| Packing quantity and moisture level | Not published |
The 100 W driver is the first cost, not the transistor
A 100 W driver at 1 GHz is a real amplifier. If it is built on the same shelf it is a 100 W part with 13 to 17 dB of gain, which means a 4 to 8 W stage in front of it, and suddenly the chain is three stages deep with two bias loops to set and two thermal paths to solve. If it is bought as a module, the module has its own lead time and its own price, and on a 1000 W transmitter the driver module is often the longest lead item.
The alternative is to divide the output. Two 500 W parts combined give 1000 W with 3 dB more drive per part in the sense that each needs 50 W at 13 dB, and a balanced pair is easier to keep stable into a mismatch. On the shelf the nearest neighbours are 750 W at 14 dB, which needs 29.9 W, and 900 W in a narrow band, and those two rows exist precisely because a 1000 W part with 10 dB of gain is a hard part to drive.
Whichever route is chosen, the gain figure has to be checked at the band edges. Ten decibels is a low gain for a 1 GHz LDMOS, and a low gain part usually means a device with a large periphery run at high current density, where gain falls fastest at the top of the band. Ask for the gain at 1215 MHz, not only at the centre.
Duty cycle is the missing figure, and it decides the heat sink
Pulse mode on a radar or IFF chain is defined by pulse width, duty cycle and peak power, and only one of the three is published. At 1 per cent duty the average output is 10 W and the average heat about 10 W, which a modest flange handles. At 10 per cent duty the average heat is 100 W. At 30 per cent it is 300 W. The same transistor needs three completely different heat sinks depending on a figure the vendor page does not carry.
That is not a small omission on a 1000 W part. It also decides the supply: 40 A average at 10 per cent duty is 4 A from the bulk capacitor bank, while at 30 per cent it is 12 A continuous, and the difference is a capacitor bank with a ripple rating rather than a bank chosen by capacitance alone. Put the energy storage close to the flange, because the pulse current loop is where a 1000 W board usually fails first.
Ask the vendor for pulse width, duty cycle, and the thermal impedance for that pulse width. A transient thermal impedance curve is the only honest way to size a pulse part, and without it the only safe assumption is the DC thermal resistance, which will oversize the heat sink and understate the junction temperature during the pulse.
The application list says ISM, the index says pulse
The vendor page lists one application, ISM RF energy, and the index marks the row Pulse. Those two do not sit well together. ISM RF energy at 1 GHz means magnetrons replaced by solid state in industrial heating and plasma, and those applications run continuous wave or at high duty with tight average power. A row rated as a pulse part with no duty figure cannot be quoted for that duty without a written statement from the vendor.
For avionics the position is the reverse and simpler. The 960 to 1215 MHz band carries interrogate and reply pulses, the duty is low, and a 1000 W peak part is exactly the right shape. A buyer with an avionics requirement should read the page's avionics sentence and ignore the ISM line, and a buyer with an ISM requirement should treat the row as unproven until the duty is in writing.
That divergence is worth flagging to the vendor directly. One sentence saying which duty the 1000 W figure is released against would settle both readings.
Where the HXC50136 sits on the shelf
The group runs 90 W, 250 W, 300 W, 750 W, 900 W narrowband and 1000 W. From 750 W to 1000 W is 1.25 dB and from 900 W to 1000 W is 0.46 dB, so the top of the ladder is crowded: three rows within 1.25 dB of each other, differing mainly in gain and in band. The 1000 W row has the lowest gain of the three and the widest band, which is the usual trade at this power.
Read as a choice, this row is for a system that needs 60 dBm from one device and can supply 100 W of drive, with a low duty cycle that keeps the average heat inside the flange. A system that cannot supply 100 W of drive takes the 750 W row at 14 dB and combines two of them if it needs the last decibel.
Before the purchase order goes out
Three procurement points follow. No orderable addendum is published for this row, so the tray quantity, the moisture level and the marking have to come from the vendor in writing before a production order is placed, and the incoming route for a 1000 W part whose parameters are only a picture is set out on the incoming inspection plan page. Buyers comparing this row against a competitor's 1000 W L band part should work from the RF power transistor cross reference, which lines the two up on gain, rail and package rather than on the peak wattage. A bill of materials that carries the 100 W driver, the combiner and the pulse capacitors can be placed through the one stop RF supply route.